During a Twitter exchange recently about the nuisance that Twitter Bots are becoming, a lot of Tweeters were asking about how to identify them and what to do about them. So I thought I might save posting 100 Tweets and give you my personal opinion and filtering process.

We should start at the beginning though. What is a bot? It is a contraction of robot, which is apt, as they work automatically or by scheduling or programming. They are very easy to create, and usually collect website links, RSS feeds or Twitter posts and then feed them to Twitter. The most common auto feeders are Twitterfeed, SocialOomph, Ping.fm, Google, Twittelator or via an API. There are many others. Luckily Twitter has made the origin of a Tweet very clear on the ‘Home’ screen of a user’s Twitter page.

This example post below has been automatically posted by Twitterfeed.

Old Oil Rig Converted Into Brand New Vacation Spot For Scuba Divers [Travel] http://bit.ly/c1sCnw
about 7 hours ago via twitterfeed

If I see a long uninterrupted list of postings like this, with its source as Twitterfeed, Socialoomph, Google or similar auto posting sources, it is probably a bot.

From this you can make a decision about continuing to follow or not. There are many very GOOD useful bots such as news channels, some great humour bots or gardening tips. It depends on your interests. Then there are RT (ReTweeting) bots that collect Tweets about a particular subject that interests people. These can include subjects like dog training, writers, new books, scuba diving or any number of interests. These bring your subject of interest into one follow.

Then there are BAD bots that are trying to sell you something. Every ten minutes. Normally with a link to a website and are repetitive in nature. If not bad, at least boring and clogging up your timeline.

Lastly are the UGLY bots that are often sex and porn sites, mail order brides, money scams and worse. I’ll leave the rest of these types to your imagination. There are also bots that send direct messages. These you will have to block as you find them.

So what can you do? When new to Twitter, the first thing you want are some followers. Not much fun with no one listening to you. In the process it is inevitable that you will end up following some bots. Usually they use a tool called ‘auto-follow back’ to make you feel good that you now have a few followers. Unfortunately, they are not good listeners. Until recently, if you unfollowed the bot, it would use a tool called ‘auto-unfollow’ to end you beautiful one way relationship. Twitter has now ended this ‘auto-unfollow’ tool, so the bot has to unfollow you manually. This means that your follower count may not go down immediately.

So you want to do a bot clean up? I have tried a number of 3rd party applications to identify and remove bots. Unfortunately they have proven to be very slow, inaccurate and unreliable. They can lead to you unfollowing many of your favourite Tweeters by mistake. The only sure fire way I have found is to use my Twitter webpage ‘Home’ page. Slow yes. But very effective.

My method is to check my ‘home’ page once a day and look through the list of recent posts. When I see a post by a suspected bot, I open the user’s Twitter page and check their last 20 posts. Then I can make an informed decision as to whether I unfollow, block or continue following.

Just one last tip. Someone setting up bots rarely sets up only one. I have found that a girl’s name followed by a number and a pretty picture a good quick bot identifier. I blocked a bot recently which used the names Lisa1234, Lisa1235, Lisa1236. Get the picture. I found 25 of these Lisa with a number bots. Not a very imaginative bot!

Well, happy bot hunting! But do be careful.

As an addition to this post, my recent post about Twitter’s Very Precious Garbage Dump may be of interest in understanding why Twitter is happy to have so many of these automated accounts. They keep the user numbers up!

Twitter Bots – The Good, Bad And Very Ugly
Tagged on:                 

5 thoughts on “Twitter Bots – The Good, Bad And Very Ugly

  • 27/09/2010 at 8:16 pm

    Cheers Derek :))

  • 27/09/2010 at 8:16 pm

    Good one. Let the war on bots begin.

  • 27/09/2010 at 8:31 pm

    Hey. This is actually kind of helpful.I'm new to twitter, so still figuring things out. Thanks :p
    (btw it's slainia)

  • 28/09/2010 at 8:17 pm

    Good info. Thanks.

  • 12/07/2011 at 1:49 am

    Thanks for explaining what bots are and how to spot some of them. Helpful to a newbie like myself.

Comments are closed.